Previous studies have shown that the facility where a patient with ESRD receives hemodialysis is a critical factor in determining his/her dialysis adequacy. This center effect is manifested by: 1) strong within center correlations in individual values of the urea reduction ratio (URR), a common measure of dialysis adequacy, and 2) wide between center variations in facility mean URR values. It is not known, however, whether there is a center effect on other clinical performance measures in hemodialysis patients such as anemia management or more importantly, mortality rates. Moreover, it is not known to what extent poorly quantified patient effects confound the measurement of center effects. In this study we will conduct a secondary analysis of existing data sets to explore the influence of center effects on anemia management (a continuous outcome like URR) and mortality rates (a binary outcome) in the hemodialysis population. We will utilize mixed effect linear models to measure the center effect on anemia management, in a regional dialysis network and, if present, determine whether the center effect is independent of dialysis adequacy. We will also use mixed effect linear models applied to US Renal Data System (USRDS) and Medicare standard analysis files (SAF) to measure the center effect influencing anemia management at the national level and show that it is independent of treatment with recombinant human erythropoietin. We include several indicators of patient socio-economic status, health and living environment to control for patient effects in the models. Finally we will adapt methods to identify and quantify the role of center effects on binary outcomes and then determine if there is a significant center effect on mortality rates in a cohort of hemodialysis obtained from the USRDS SAF.